ActaBot. Croat. 59(1), 101-109,2000

CODEN: ABCRA25 ISSN 0365-0588 UDC 581.5(620)

Dedicated to Prof. dr. LJUDEVIT lUJANlC on the occasion o f his 70th birthday.

Some observations on the plant communities of Dungul Oasis (Western Desert, Egypt) R einhard B ornkamm 1*, I rina S pringuel 2, F rank D arius 1, M ohamed G abbr S heded 2, M agdi R adi2 1 Technical University of Berlin, Institute of Ecology and Biology, Rothenburgstr. 12, D-12165 Berlin, Germany 2 University of the South Valley, Unit of Environmental Studies and Devel­ opment, 81528 Aswan, Egypt During a visit to the Dungul Area (Western Desert, SW Egypt), several vegeta­ tion records were made. The number of species per plot (25 m2) was extremely low: mainly monospecific stands were found. A combination of two (or even three) species was reached either in the transition belts of herbaceous species or of one tree species with understory plants. Special attention was paid to the rare palm species Medemia argun, endemic to Nubia. A famous old specimen known from literature was encountered in dead state, but several living young specimens were observed. Key words: Sahara, desert, oasis, vegetation, palm, Medemia argun, Nubia, Egypt

Introduction In the southern part of the Western Desert of Egypt the Nubian Tableland de­ scends to the East in the direction of the Nile Valley near Aswan and Lake Nasser, with a sharp and spectacular escarpment (»Sinn El-Kaddab«) to the Lower Nubian Plain (B utzer 1965). Within this escarpment, two small, unin­ habited oases are situated: Kurkur and Dungul (Fig. 1). In both oases, geological, archaeological and botanical studies have been carried out, but Wadi Kurkur, closer to Aswan (approximately 60 km) has received much more research atten­ tion than Dungul (approximately 160 km). As examples, we would like to quote here the following important investigations (where further relevant literature is cited): S hata (1962), R eed (1964), B utzer (1965), B oulos (1966). and S heded and H assan (1998) for Kurkur; Z ahran (1968), and B oulos (1968) for Dungul.

* Corresponding author: E-mail: [email protected] ACTA BOT. CROAT. 59(1), 2000

101

BORNKAMM R., SPRINGUEL I.. DARIUS F „ SHEDED M. G., RADI M.

The Dungul Oasis complex consists of two parts: (1) Dineigil Oasis and (2) Dungul Oasis proper (i.e. Dungul South and Dungul North sensu Z ahran 1968, fig. 2). Dineigil is located at the very edge of the escarpment in a high position at 23 °N, 24°6’E, Dungul in the Wadi Dungul already inside the Nubian Tableland (23°26’N, 31°37.3’ E) in a lower position (250 m, U hden 1930). The geological conditions are described by H endriks et al. 1987 (Fig. 11, p. 77), and a lithostratigraphic section is given by E l S hazly et al. (1977, plate X X IV ). Both Dineigil and Dungul receive their water as a result of the blockage of drainage lines of an artesian aquifer(EL S hazly et al. 1977). Soil analyses carried out by Z ahran (1968) showed that soils were saline (> 0.5% Cl) under Tamarix nilotica, and in the topsoil under Imperata cylindrica and Juncus rigidus, but practically not saline at all (mostly < 0.1% Cl) under Stipagrostis uulnerans and Alhagi graecorum, and in the deeper soil layers under Imperata cylindrica and Juncus rigidus. Little information was available until Z ahran (1968) gave a description of the vegetation. He found in Wadi Dungul the community types of Salsola imbricata, Tamarix amplexicaulis, Tamarix aphylla, and Stipagrostis uulnerans, in Dungul Oasis the community type of Imperata cylindrica, and in Dineigil Oasis the community types of Alhagi graecorum, Juncus rigidus, and Imperata cylindrica. The highlight of the floristic characters is the occurrence of the palm species Medemia argun DC. which grows together with the date palm (Phoenix dactylifera) and the doum palm (Hyphaene thebaica), and is known in Egypt out­ side Dungul only from the nearby Nakheila Oasis (B oulos 1968). It was, how­ ever, an important tree in ancient Egypt (TAckholm and D rar 1950; B oulos 1968; ScHOSKEet al. 1992). During a recent visit to the Dungul oases (December 6-8, 1998) we tried to get a first impression as to which combinations of species occurred, and especially which understory plants were combined with the three palm species. 102

ACTA BOT. CROAT. 59(1). 2000

PLANT COMMUNITIES OF DUNGUL OASIS

Methods For vegetation analysis a standard plot size of 25 m2 was used. As in former investigations (B ornkamm and K ehl 1990) the shape of the plots could vary ac­ cording to the shape of the stand, but not the size. This made a direct comparison of the number of species per plot feasible. In each plot the cover (%) of dry and green parts of every species was recorded, likewise the height (dm) of the green parts. The species names follow B oulos (1995).

Results and Discussion The main part of the vegetated area is open and is dominated by either Juncus rigidus or Alhagi graecorum. The records of Dineigil are presented in Table 1. In a few cases, in burnt areas, Imperata cylindrica is the most frequent although not the dominant species at present, because above-ground only stubble has survived the disturbance (rec. 1^1). According to the ordinary zonation around wells (K ehl 1987) Juncus rigidus grows closest to the water holes, whereas here in an Alhagi stand a recently dug water hole shows a seepage of more than 3 m below surface! Four tree species were observed. Phoenix dactylifera was found just becom­ ing established in an Alhagi stand, but a mature date grove showed Imperata as understory species (Fig. 2). Acacia raddiana was accompanied by Alhagi, Aca­ cia ehrenbergiana by Alhagi and Imperata. Hyphaene was found in the juvenile stage in a Juncus stand. The large grove of Hyphaene thebaica at the Ain El-Gaw spring showed Sporobolus spicatus in the ground layer. Here it was already men­ tioned by Z ahran (1968).

Tab. 1. Records from Dungul Oasis. R = number of records; C = total cover (%); N = number of species; 1 = cover of living, green parts (%): d = cover of dead parts (%); h = height (dm).

Imperata cylindrica

Hyphaene thebaica

Medemia organ

Phoenix dactylifera

Tamarix aphylla

Tamarix niiotica

R C N 1 d h 1 d h 1 d h 1 d h 1 fl h 1 d h

1 60 1 30 40 12

2 3 4 5 6 7 8 9 60 22 70 100 90 80 80 90 1 1 1 1 2 1 2 2 20 2 20 20 3 3 40 20 50 30 3 5 6 8 7 8 6 8 50 30 50 50 40 30 100 120 80 80 90 20 10 80 100

ACTA BOT. CROAT. 59(1), 2000

10 11 12 13 14 15 90 100 100 100 100 100 2 2 2 1 1 2 1 2 1 5 3 2 7

75 80 25 20 100 110 70 50 40 100 100 70 10 20 22 60 60 35 103

BORNKAMM R., SPRINGUEL I., DARIUS F., SHEDED M. G., RADI M.

Fig. 2. Phoenix dactylifera and lmperata cylindrica in Dineigil Oasis

In Dungul we visited Dungul Oasis but not the other parts of the very ex­ tended Wadi Dungul. Our records are presented in Table 2. In the Dungul Oasis, lmperata cylindrica was the dominant species. Here too, in some burned areas it was still the most frequent species without being dominant. Three tree species were observed, all of them palms. As far as a ground layer was developed it was made up exclusively of lmperata cylindrica (Fig. 3). The eastern end of the vege­ tated area was made up of a very dense scrub of several Tamarix species devoid of any herbaceous plants. It seemed that vegetation had not changed very much since the investigation of Z ahran (1968). The Argun Palm needs special comment. The species Medemia argun was described by P. G. V on W ürttemberg in W endland (1881). In taxonomic liter­ ature it is sometimes regarded as a member of the genus Hyphaene: Hyphaene argun (J ackson 1893, 1894). The similarity is, indeed, striking. The species was detected in Dungul by Boulos, Täckholm and Zahran in November 1963, and in 104

ACTA BOT. CROAT. 59(1), 2000

PLANT COMMUNITIES OF DUNGUL OASIS

Tab. 2. Records of Dineigil Oasis. Abbreviations as in Table 1.

hmcus rigidus

Imperata cylindrico

Alhagi gra eco m

Phoenix dactylifera

Acacia raddiana

Acacia

R 1 C 100 N 1 1 90 d 10 h 16 1 d h 1 d h 1 d h 1 d h 1

2 90 2 50 20 14

5 5 6

3 90 2 45 25 16

5 5 6

4 90 2 90 20

5 70 2 25 10 12

35 5 6

6 17 2

7 50 3

8 40 2

15 0

5 8 20 20 8 3 8 2

10 10 7 10 10 2

1 1 2 5 5 20

9 90 1

75 15 12

10 10 1

5 5 4

11 12 13 80 100 100 2 2 3

3 10 8 0.5 0.5

15 15 3

1

2

14 90 3

15 80 2

16 90 2

20 20 8

5 0 2 1 1.5

80 150 70 100 45

80 100

80

70

55

ehrenbergiana

Hyphaene thebaica

Sporobolus spicatus

d h 1 d h I d h

80 100 15 5 5

Fig. 3. Two groups of Hyphaene thebaica with Imperata cylindrica.

Nakheila by Issawy in December 1964 ( B o u l o s 1968). In 1963 one large tree was developed, surrounded by seven »baby palms« ( Z a h r a n 1968). During our visit we counted 37 shoots. One of them, apparently the mother tree, was dead ACTA BOT. CROAT. 59(1), 2000

105

B o r n k a m m R., S p r in g u e l I., D a r iu s F „ S h e d e d M . G., R a d i M.

and broken (Fig. 4); it had reached a height of ca 10 m. Seven other shoots at­ tained a height of > 3 m. Among the largest specimens of them (ca 8 m) were 1 fe­ male and 3 males; the remaining 29 shoots were only ca 1-2 m high. As also re­ ported by B oulos (1968), an enormous number of fruits were lying on the ground.

Fig. 4. The remnants of the old Medemia tree and younger specimens in Dungul Oasis.

The area between the two oases is not completely void of vegetation. Salsola imbricata occurred at favourable sites in small wadis or runnels, and in depres­ sions (Tab. 3). Eventually other species like Phoenix dactylifera or Fagonia indica may occur. This scarce vegetation seems to be typical for the Dungul-Kurkur area at large (S heded and H assan 1998). Most of the species observed are able to build up monospecific stands. Vege­ tation types with less than two species on the average, where the dominant spe­ cies is not regularly accompanied by at least one additional species, are called »stands« according to B ornkamm and K ehl (1990); they are not plant »commu­ nities«. In our case this applies to 19% of the records from Dineigil, to 47% of the records from Dungul, and to 67% of the records from the area between the oases. The average species number per plot is 2.1 in Dineigil; 1.5 in Dungul; and only 1.3 in the area in between. A higher number of species (in our records not more than 2 or 3) can be reached in two ways: (1) In the herbaceous vegetation fre­ quently a transition belt is developed between two monospecific stands. (2) In the woody vegetation frequently a tree layer and a field layer are developed, thus creating a savanna-like structure. Further studies are needed to find out how far the performance of species (e.g. cover, height, vitality) in the mixed stands dif­ fers from that in the monospecific stands. 106

ACTA BOT. CROAT. 59(1). 2000

PLANT COMMUNITIES OF DUNGUL OASIS

Tab. 3. Records from the area between the oases. Abbreviations as in Table 1. r c N 1 d h 1 d h 1 d h

Phoenix daclylifera

Salsola imbricóla

Fagonia indica

1 100 1 80

2 60 1

3 95 1

4 70 1

5 7 2

6 10 2

60 10 10

95

40 30 7

0.1 5 7

10 4

7 2 1

8 0.5 1

2 2

0.5 2

150

15

Already the earlier investigations in small, uninhabited oases of the Western Desert led to the result that every one of them has its peculiar traits (B ornkamm 1986). Great differences between Dineigil and Dungul were found, with Acacia species and relatively largeAlhagi stands occurring in Dineigil (Tabs. 1,2), whe­ reas in Dungul in addition to palm groves stands of Tamarix woodland are devel­ oped, and larger areas are covered by Imperata cylindrica (Fig. 3). This means that-according to the results of the soil analyses by Z ahran (1968)-Dungul exhibits a more saline character than Dineigil. Quantitatively, the difference between Dineigil and Dungul can be elucidated by computation of the similarity index (S 0 rensen 1948) of the species lists of the oases. In Table 4 similarity values are presented for Dungul (species list from the present paper in­ cluded additional species found by Z ahran (1968), Dineigil (present paper), Kurkur (S heded and H assan 1998) and the two neighbouring oases in the West­ ern Desert, Nakhlai and Takhlis (E l H adidi 1980; B ornkamm 1986). Taking into account all species the Sprensen value for Dineigil/Dungul is as low as 33%. Dineigil is more similar to Kurkur than to Dungul, while Dungul is more similar to Nakhlai and Takhlis than to Dineigil. This means that we have a clear W -E gradient, which is underlined by the total species numbers, which range from Takhlis and Nakhlai in the West (4 species, comprising 2-3 woody species) to Kurkur in the East (20 species, comprising 6 woody ones). This gradient may be due to the increase of the incidence of rainfall from W to E. Taking into account the woody species only, higher S0rensen values were achieved. Here the similar­ ity between Dineigil and Kurkur amounts to 80%, between Dineigil and Dungul to only 40%. Otherwise no clear differences exist which is probably due to the extremely low number of species involved.

Tab. 4. Floristic similarities according to S o r e n s e n (1948) between the Dungul, Dineigil, Kurkur, Nakhlai and Takhlis Oases (in%). Upper half: all species; lower half: woody species only. N = number of species. Kurkur Kurkur Dineigil Dungul Nakhlai Takhlis

X

80 67 50 67 6

Dineigil 43 X

40 33 57 4

ACTA BOT. CROAT. 59(1), 2000

X

Nakhlai 17 17 43

50 50 6

33 2

Dungul 31 33

X

Takhlis 25 33 43 50 X

N 20

10 8 4 4

3 107

B o r n k a m m R „ S p r in g u e l I., D a r iu s F „ S h e d e d M . G „ R a d i M.

The present state of the vegetation is influenced by human activity. As long ago as 1930 U hden mentioned several groups of doum palms and some feeble date palms which suffered from charcoal-burning by the Bedouins. Nevertheless H offman (1979, p. 55, Fig. 14) regarded the vegetation of Dungul as »one of the last remnants of the vegetation that covered Sahara during pluvial periods«. This evaluation stresses the historical, archaeological and ecological importance of the Dungul area. The ecosystems here are rare, to some extent even unique, vul­ nerable and endangered-typical traits that make protection necessary. In the same way as the need for protection has been expressed for parts of the Eastern Desert (S pringuel 1997), the Dungul-Kurkur area is also very well worth being protected, also at a national scale. A firm basis for the understanding of the eco­ systems involved during the process of protection should be obtained by careful studies in the near future

References BORNKAMM, R., 1986: Flora and vegetation of some small oases in S-Egypt.

Phytocoenologia 14, 275-284. BORNKAMM, R„ KEHL, H., 1990: The plant communities of the Western Desert

of Egypt. Phytocoenologia 19, 149-231. BOULOS, L., 1966: A natural history study of Kurkur Oasis, Libyan Desert,

Egypt. IV. The Vegetation. Postilla Peabody Museum of Natural History, Yale University (New Haven, Connecticut, USA) 100, 1-22. BOULOS, L., 1968: The discovery of Medemia Palm in the Nubian Desert of Egypt. Bot. Not. 121, 117-120. BOULOS, L., 1995: Flora of Egypt - checklist. A1 Hadara Publ., Cairo. BUTZER, K. W., 1945: Desert landforms at Kurkur Oasis, Egypt. Ann. Ass. Amer. Geogr. 55, 578-591. EL H a d id i , M. N., 1980: Vegetation of the Nubian Desert (Nabta Region). In: WENDORF, F., SCHILD, R. (eds.), Prehistory of the Eastern Sahara. Studies in Archaeology, Appendix 5, 345-351. Academic Press, New York. El S h azt.y, e . m „ Abdel H ady, M. A., El Kassas , A., E l Shazly, M. M., ABDEL MEGID, A. A., T am er , M. A., 1977: Geology and groundwater conditions of Tushka basin area, Egypt, utilizing Landsat satellite images. Remote sensing center, Academy of Scientific Research and Technology, Cairo. HENDRIKS, S. F., LUGER, P„ BOWITZ L, KALLENBACH H., 1987: Evolution of

the depositional environments of SE-Egypt during the cretaceous and lower tertiary. Berliner Geowiss. Abh. (A) 75, 49-82. JACKSON, B. D., 1893: Hyphaene. Index Kewensis, part II, 1195. Clarendon

Press, Oxford. JACKSON, B. D., 1894: Medemia. Index Kewensis, part III: 182. Clarendon Press,

Oxford. 108

ACTA BOT. CROAT. 59(1), 2000

PLANT COMMUNITIES OF DUNGUL OASIS

K e h l , H., 1987: Zonation and establishment of vegetation in selected uninhab­

ited Egyptian and Sudanese oases. Catena 14, 275-290. REED, C. A., 1964: A natural history study of Kurkur Oasis, Libyan Desert, Western Governate, Egypt. I. Introduction. Postilla Peabody Museum of Nat­ ural History, Yale University (New Haven, Connecticut, USA) 84, 1-20. SCHOSKE, S., K r e ib l , B., G e r m e r , R., 1992: »ANCH«. Blumen für das Leben. Pflanzen im alten Ägypten. Staatl. Samml. Ägyptischer Kunst, München. S h a t a , A., 1962: Remarks on the geomorphology, pedology and ground water potentialities of the southern entrance to the New Valley. Part 1. The Lower Nubia area, Egypt, U.A.R. Bull. Soc. Géogr. Egypt 35, 273-299. SHEDED, M. G., H a s s a n . L. M., 1998: Vegetation of Kurkur Oasis in SW Egypt. J. Union Arab Biol., Cairo 6 B, Botany, 129-144. S0RENSEN, T., 1948: A method of establishing groups of equal amplitude in plant sociology based on similarity of species content. Biol. Skr. K. danske Vidensk. Selsk. 5, 1-34. SPRINGUEL, L, 1997: Vegetation, land use and conservation in the SE Desert of Egypt. In: BARAKAT, H. N., HEGAZY, A. K. (eds.), Reviews in ecology:

Desert conservation and development, 177-206. Metropole, Cairo. TÄCKHOLM, V, DRAR, M., 1950: Flora of Egypt 2. Bull. Fac. Sei. Cairo Univ.

28, 296-302. UHDEN, R., 1930: Reise von Deböt in Unternubien nach den Oasen von Kurkur und Dungul. Petermanns Mitt. aus J. Perthes Geogr. Anstalt 76, 184-188. WENDLAND, H., 1881: Beiträge zu den Borassineen. Botanische Zeitung 39,

91-96. ZAHRAN, M. A., 1968: Ecological study of Wadi Dungul. Bull. Inst. Désert

Egypte 16: 127-143.

ACTA BOT. CROAT. 59(1 ), 2000

109